Main Page

From Harlem

Jump to: navigation, search


[edit] HARLEM - Molecular Modeling Package


HARLEM is a multipurpose multi-scale molecular simulation package with unique capabilities not found in other programs. It also provides convenient graphical interface to a few popular molecular modeling packages such as AMBER and GAUSSIAN.

HARLEM ( abbreviation stands for HAmiltonians to Research LargE Molecules) is developed by Igor Kurnikov while he was a graduate student and a postdoc at the University of Pittsburgh and later at Northwestern University. The first application within HARLEM was developed to study long range electron transfer reactions in biological molecular systems and in nanoelectronic devices. Gradually the scope of the program was extended to incorporate a variety of molecular modeling techniques and approaches. It has been used to study such diverse biological problems as oxidative stress in biology and ionic currents through membrane protein channels.

HARLEM is being actively developed and maintained by Igor Kurnikov and Maria Kurnikova in Maria kurnikova's research group at Carnegie Mellon University in Pittsburgh, PA (USA). Other people who contribute to development are HARLEM are Nikolai Simakov, Jose FLores Canales, Kirill Speranskiy.

[edit] Getting Started

[edit] Download and Install HARLEM on Windows

1. Download

2. Unpack content of file into c: directory, this should create directory c:\harlem

  * c:\harlem contains an executable harlem.exe and several other files files 
  * Subdirectory c:\harlem\residues_db with residues and force-field databases (currently files amber_94_ff.dat, res_XXX.hlm files)
  * Subdirectory c:\harlem\scripts: with Python scripts providing access to HARLEM classes
  * Subdirectory c:\harlem\basis:   with Gaussian Quantum Chemical basis sets in DALTON format
  * Subdirectory c:\harlem\doc:     containing HARLEM documentation.
  * Subdirectory c:\harlem\examples with a few input files and scripts to run HARLEM jobs

3. ONLY If installation has been performed into the directory other than c: (for example d:\) please set environment variable HARLEM_HOME to this directory (SET HARLEM_HOME=d:\)

4. For first time HARLEM installation in c:\harlem execute:

  • vc10sp1_redist_x86.exe
  • w_fcompxe_redist_ia32_2013.1.119.msi

This will install DLL redistribution packages of Microsoft c++ v10 and Intel Fortran v13

this completes installation of HARLEM to execute on a single core processor

[edit] To execute HARLEM on Windows computer

Option A: Double click on the file harlem.exe in c:\harlem directory.

The program will start with two windows:

  • the main working window with the main menu on top
  • a console window for printing out diagnostic, output information, and some random error messages to be ignored.

it is a good idea to make shortcut link of harlem.exe to desktop

  * in windows 7 to make file extensions (such as .exe or .pdf etc.) visible: in any open folder in main menu chose 
  organize->"Folder and Search options"->tab "View": uncheck "Hide extensions of known file types"

Option B: In your working directory chose a molecular structure file molec.pdb or molec.hlm (native harlem format) and double click on it to start harlem (after it was associated with the proper file-type). This is preferable because this will set the current working directory to the directory where the initial structure resides.

[edit] To install HARLEM on multiple cores in parallel mode on Windows computer

5. in c://HARLEM/ execute mpich2-1.2-win-ia32.msi

this installs the MPICH library of parallel functions

[edit] To execute in parallel on Windows (Required to e.g. run TI)

option 1:

1. In the working directory have the configurational for wmpiexec.exe file "myname.mpi"

(can be copied from c:\harlem\examples\MD\TI_MG_SOLV)
See file content

option 2:

1. run wmpiexec.exe (from MPICH2) and set up "job" to create and save this file ("myname.mpi")

2. Run wmpiexec.exe by double clicking on the file myname.mpi (or run from its location in Program Files/MPICH2)

3. Load Job (chose the same file myname.mpi)

4. Execute

this should start HARLEM, which is now in parallel mode on the number of cores equal n (second line in the file myname.mpi)

[edit] HARLEM Documentation

Reasonably up-to-Date HARLEM documentation is distributed with harlem and can be found in c:\harlem\doc

  • C:/harlem/doc/harlem_prog_manual.chm - in windows documentation format
  • C:/harlem/doc/prog_man_html/index.html - in html format

elsewehere on the internet a set of older documents can be found in

 Beware! currently these pages are old (2004-2006) - 
 they are relevant for the main architecture of the program 
 but there are lots of new features and some discrepancies in class structure etc.

[edit] Bare Essentials to Get Started

1. c:\harlem.exe - is the main executable to run

  running it opens two windows: main window, which is working window, and console window for program output only.

2. main menu: File->open molecule opens a dialogue window to open pre-existing molecular structure. By default the program expects HARLEM native file format: hlm. Chosing a different file format will show file names with that file format. A common structure file format is PDB (protein data bank file format)

3. To see what kind of molecule is currently loaded:

main menu: Edit-> Edit residue parameters - opens a table witht he list of residues (mers of the polymer or individual molecules) with their names, numbers and some other properties listed.

4. To see the complete list of atoms:

main menu: Edit-> Edit atom parameters - opens a table with he list of all selected atoms with their names, numbers and some other properties listed. Chose to see, mass, coordinates, force field names, charges and other properties depending on the model that is activated.

5. The program is designed to perform molecular modeling and visualize molecules, therefore it operates with known templates for molecular residues.

The complete list of known residues can be seen via Main menu: Edit-> Show Residue DB.
This also pops up a new molecular view window with all residues at once. If one of the residues is selected from the list, it will be shown on screen alone.

6. To save a molecule in the current view window (termed MolSet)

main menu: File-> Save Molecule as, then chose the format and the directory

7. All commands and pop up dialogue windows operate on a currently selected subset of atoms. If selections has changed, press "update atom list" in Edit Atom Parameters window, or "update residue list" in Edit residue parameters. ALso, the window information will be updated after a window is closed and open again.

8. Atom naming conventions and Selection of atom groups

9. Coloring schemes:

  • the default coloring scheme for atoms is called cpk, which is
    • carbon - grey, oxigen - red, nitrogen - blue, hydrogen - white, sulphur - yellow
  • to color selected atom to other colors:
type in command line in main window next to "Execute command": color color_choice
additional color_choices are: cpk, orange, green, purple, cyan, etc..
or use the RGB scheme, e.g. color [100,100,100] - for grey

10. Molecular and atomic representation

main menu: Display-> Display mode lists a number of atom representations and protein structure representations - just try them
main menu: Display-> colors lists choices for coloring schemes that recognize molecular structure of proteins
  • some of the representations can be used on the command line, e.g. ribbons, or trace with parameters to control width

 Tutorial below on Molecular Visualisation is the best way to get started and learn basic functionality of Harlem Graphical Interface

[edit] HARLEM Tutorials

Download and unzip the file. Follow the instructions in the file harlemtutorial.doc

  • Molecular Dynamics
    • Tutorial 2: Getting started with Molecular Dynamics and python scripting
  • Monte Carlo method
    • Tutorial 3: A simple tutorial for Monte Carlo simulation and python scripting

[edit] Useful Python Scripts for HARLEM

[edit] For Developers

[edit] Projects

  1. HIV-RT Project
Personal tools